OSINT & SURVEILLANCE
Intelligence cycle · Domain recon · Shodan · Google dorking · SOCMINT · Geospatial · Maltego · Dark web · Metadata · Investigator OPSEC
OSINT Fundamentals & Intelligence Cycle
- OSINT uses only publicly available data - passive collection is always legal, active probing requires authorisation.
- The intelligence cycle (Plan, Collect, Process, Analyse, Disseminate) prevents wasted effort and keeps investigations focused.
- OSINT covers five major target categories: people, organisations, infrastructure, geospatial, and dark web.
- Investigator OPSEC requires layered defences: dedicated VMs, VPNs, Tor, and sock puppet accounts - never research from your personal identity.
- Sock puppets need 30+ days of aging with realistic posting patterns before they are usable for sensitive research.
Think of the mama mboga at the estate gate. She never steps inside your house, yet from her stall she learns your routine — when you leave for work, who visits on weekends, that you have started buying baby food. She pieced it all together from what simply passed by in plain view. OSINT is that same craft done on purpose: building a full picture of a target purely from what is already out in the open.
► Intelligence Disciplines: OSINT in Context
Intelligence collection is categorised by source type. Understanding where OSINT sits within the broader intelligence landscape clarifies both its power and its limits.
| DISCIPLINE | FULL NAME | SOURCE | OSINT ROLE |
|---|---|---|---|
| OSINT | Open Source Intelligence | Publicly available: web, social, records, media | Primary discipline covered here |
| HUMINT | Human Intelligence | Human sources, informants, interviews | Social engineering overlaps here |
| SIGINT | Signals Intelligence | Intercepted communications, RF signals | Govt/military domain, some WiFi overlap |
| IMINT | Imagery Intelligence | Satellite, aerial photography | Commercial sat imagery is now OSINT |
| GEOINT | Geospatial Intelligence | Location data, mapping, terrain analysis | Flight/ship tracking, geo-verification |
| TECHINT | Technical Intelligence | Analysis of foreign technology/equipment | Malware analysis, firmware analysis |
► The Intelligence Cycle
Professional intelligence work follows a structured cycle. This prevents wasted effort, keeps investigations focused, and ensures findings are usable. Skipping planning is the most common beginner mistake.
PHASE 1: PLANNING & DIRECTION
- Define the intelligence requirement (PIR - Priority Intelligence Requirement)
- Scope: What exactly do you need to know?
- Target: Organisation, individual, infrastructure, or combination?
- Legal check: Do you have authorisation for active recon?
- Output format: Report, briefing, graph, timeline?
|
v
PHASE 2: COLLECTION
- Passive: Reading public data (WHOIS, social media, news)
- Semi-passive: Normal user-level traffic (visiting their homepage once)
- Active: Direct probing (scanning, crawling) - requires authorisation
- Sources: Web, social media, DNS, leaked databases, satellite imagery
|
v
PHASE 3: PROCESSING
- Normalise and structure raw data
- Deduplicate: remove redundant hits
- Correlate: email -> username -> social profile -> IP
- Filter noise: distinguish signal from background
|
v
PHASE 4: ANALYSIS
- Pattern recognition: behaviours, timelines, relationships
- Network mapping: who connects to whom?
- Gap analysis: what is missing and why?
- Confidence levels: high / medium / low confidence per finding
|
v
PHASE 5: DISSEMINATION
- Structured report with citations for every claim
- Visualisations: link graphs, timelines, maps
- Executive summary + technical appendix
- Recommendations for follow-up► Passive vs Active Reconnaissance
This distinction is legally and operationally critical. Passive recon means you never send a packet to the target. Active recon means the target's systems see your traffic.
| TYPE | CONTACT WITH TARGET? | EXAMPLE | DETECTABILITY | LEGAL RISK |
|---|---|---|---|---|
| Passive | None - third-party data only | Reading WHOIS, searching Google, checking crt.sh | Zero | None - always legal |
| Semi-passive | Normal user-level traffic | Visiting their public homepage, LinkedIn | Minimal - normal web traffic | None |
| Active | Direct probing of target systems | Port scanning, banner grabbing, dir fuzzing | High - appears in server logs | Illegal without written authorisation |
► OSINT Target Categories
| CATEGORY | WHAT TO COLLECT | PRIMARY SOURCES |
|---|---|---|
| People (individuals) | Full name, email, phone, username, social profiles, location, relationships | Social media, data brokers, breach databases, people search engines |
| Organisations | Domains, IPs, employees, org structure, tech stack, locations, subsidiaries | WHOIS, LinkedIn, Shodan, job postings, Companies House / SEC filings |
| Infrastructure | Open ports, running services, SSL certs, cloud assets, network ranges | Shodan, Censys, crt.sh, BGP databases, cloud storage scanners |
| Geospatial | Physical locations, building layouts, movement patterns, vehicle routes | Satellite imagery, Google Maps, flight/ship tracking, social media geotagging |
| Dark web | Leaked credentials, breach data, threat actor chatter, stolen data | Tor browser, paste sites, dark web search engines, breach monitoring services |
► Legal Framework
OSINT practitioners must understand the legal boundaries of their work. The law differs by jurisdiction but two frameworks are most relevant globally.
COMPUTER FRAUD AND ABUSE ACT (CFAA) - United States - Prohibits accessing computer systems without authorisation - "Without authorisation" includes exceeding permitted access - Passive OSINT (reading public data) is not covered - Active scanning of systems you do not own = potential violation - Civil and criminal liability GDPR - European Union (and UK equivalent) - Collecting personal data requires a lawful basis (Art. 6) - Legitimate interest can justify OSINT in professional contexts - Storing and processing EU personal data has strict requirements - Data minimisation: collect only what is necessary - Right to erasure applies even to OSINT-collected data COMPUTER MISUSE ACT - United Kingdom - Section 1: Unauthorised access to computer material (up to 2 years) - Section 3: Unauthorised acts with intent to impair systems (up to 10 years) - Applies to any UK-connected system regardless of your location PRACTICAL RULE: If a human could manually read the data by visiting a public webpage without logging in, it is almost always legal to collect. If it requires authentication, bypassing controls, or automated mass collection of personal data, get legal advice first.
► OPSEC for Investigators
Investigators can expose themselves while conducting research. LinkedIn notifies people when you view their profile. Websites log your IP. Google tracks search history. OPSEC is not paranoia - it is professionalism.
LAYER 1: Dedicated browser profile Firefox Multi-Account Containers - one container per target Never mix research with personal browsing LAYER 2: VPN for semi-passive research Paid, no-logs VPN (Mullvad, ProtonVPN) Never free VPNs - they sell your data LAYER 3: Tor for visiting target websites torsocks curl https://TARGET_SITE Changes exit IP on each circuit LAYER 4: Sock puppet accounts Separate email + phone per research persona Created from VPN/Tor, never from personal IP Age accounts 30-60 days before use LAYER 5: Isolated VM per investigation Whonix, Tails, or TraceLabs OSINT VM Revert snapshot after each investigation LAYER 6: Strip metadata before sharing exiftool -all= screenshot.png
► Sock Puppet Account Methodology
Sock puppets are research personas used to access platforms without revealing your identity. Creating believable sock puppets is an art that requires patience and consistency.
1. PERSONA PLANNING - Choose a realistic name (use name generators + census data) - Pick age, location, profession consistent with target audience - Design backstory: where did they study? What do they like? - Prepare profile photos (AI-generated faces: thispersondoesnotexist.com) 2. ACCOUNT INFRASTRUCTURE - Dedicated email via ProtonMail or Tutanota (from VPN/Tor) - Prepaid SIM or virtual number (MySudo, Google Voice) for 2FA - Create from a VPN IP you do not use for anything personal 3. ACCOUNT AGING (minimum 30 days before use) - Post generic content: news shares, hobby comments - Follow public figures, join communities relevant to persona - Build realistic follower/following ratio - Vary posting times to match persona's timezone 4. OPERATIONAL SECURITY - Never access sock puppet from personal IP - Never link to your real identity in any way - Use different browsers per persona - Log out and clear cookies between sessions 5. RETIREMENT - When investigation ends, stop using the account - Do not delete immediately - dormant accounts look natural - Delete after 6 months of inactivity
► OSINT Tools Landscape
| CATEGORY | FREE TOOLS | COMMERCIAL TOOLS |
|---|---|---|
| Domain recon | whois, dig, amass, subfinder, crt.sh | DomainTools, SecurityTrails, PassiveTotal |
| Internet scanning | Shodan (free tier), ZoomEye (free) | Shodan Enterprise, Censys Teams, BinaryEdge |
| Social media | Sherlock, Maigret, Twint, Osintgram | Pipl, Maltego commercial, SocialLinks |
| Email OSINT | Hunter.io (free tier), HaveIBeenPwned | Dehashed, Snusbase, IntelligenceX |
| Link analysis | Maltego CE (limited), Spiderfoot | Maltego Pro/Enterprise, Spiderfoot HX |
| Automated | Recon-ng, theHarvester, Spiderfoot | Recorded Future, DarkOwl, Flashpoint |
- OSINT uses only publicly available data - passive collection is always legal, active probing requires authorisation.
- The intelligence cycle (Plan, Collect, Process, Analyse, Disseminate) prevents wasted effort and keeps investigations focused.
- OSINT covers five major target categories: people, organisations, infrastructure, geospatial, and dark web.
- Investigator OPSEC requires layered defences: dedicated VMs, VPNs, Tor, and sock puppet accounts - never research from your personal identity.
- Sock puppets need 30+ days of aging with realistic posting patterns before they are usable for sensitive research.